Method for compensating nonlinear coulomb friction of linear motor

Abstract

The application discloses a linear motor nonlinear coulomb friction force compensation method and belongs to the field of linear motor friction force compensation. max (j) and the position error peak value e min (j) at the speed reverse zero-crossing point, when e max (j) exceeds the positive zero-crossing point position peak reference value e r or e min (j) exceeds the reverse zero-crossing point position peak reference value -e r , according to the result of e r -e max (j) or -e r -e min (j), a corresponding feedback controller is used to control the output of a compensation square wave amplitude limit value for coulomb friction force compensation as a feedforward compensation, and the feedforward compensation is fed back to a current instruction to complete the compensation of nonlinear coulomb friction force. The application is used for the compensation of nonlinear coulomb friction force in the linear motor motion process.

Description

TECHNICAL FIELD

[0001] The present application relates to a linear motor nonlinear coulomb friction compensation method, and belongs to the field of linear motor friction compensation. BACKGROUND

[0002] The linear motor of "zero transmission" is directly connected with the load, and while realizing direct transmission, it inevitably brings about the problems of positioning force, low-speed friction and nonlinear electromagnetic thrust, etc., so that the motor itself is more easily affected by load disturbance.

[0003] In order to meet the high performance requirements of the linear motor, disturbance suppression becomes a non-negligible problem. Among various disturbances, the adverse effects of friction on servo motion control system mainly include low-speed crawling, large steady-state static error and limit cycle oscillation problem, which limits the further improvement of the performance of the linear motor. Figure 3 As shown in the figure, the linear motor speed and position error waveform are collected by experiment. It can be seen that, in the zero-crossing stage, due to the sudden change of the direction of coulomb friction, a position error peak value will be caused. Therefore, it is very necessary to compensate for the nonlinear friction at the speed zero-crossing stage. SUMMARY

[0004] In view of the problem of excessive position error peak value caused by the change of coulomb friction direction at the speed zero-crossing stage of the linear motor, the present application provides a linear motor nonlinear coulomb friction compensation method.

[0005] The linear motor nonlinear coulomb friction compensation method of the present application comprises,

[0006] The position error peak value e max (j) at the speed forward zero-crossing point in each motion cycle of the linear motor is obtained by sampling the self-learning feedforward compensator, min (j) at the speed reverse zero-crossing point, max (j) exceeds the forward zero-crossing point position peak value reference e r (j) or the reverse zero-crossing point position error peak value e min (j) exceeds the reverse zero-crossing point position peak value reference-e r (j), according to the result of e r -e max (j) or-e r -e min (j), the corresponding feedback controller is used to control the output of the compensation square wave amplitude limit value for coulomb friction compensation as feedforward compensation, and the feedforward is to the current command to complete the compensation of nonlinear coulomb friction, wherein j represents the motion cycle.

[0007] According to the linear motor nonlinear Coulomb friction compensation method of the application, when the positive zero-crossing position error peak value e max (j) is within the positive zero-crossing position peak reference value e r , and the negative zero-crossing position error peak value e min is within the negative zero-crossing position peak reference value -e r , the feedback controller is turned off to keep the compensation square wave as the feedforward compensation at the current amplitude, and the feedforward to the current command is completed to compensate the nonlinear Coulomb friction.

[0008] According to the linear motor nonlinear Coulomb friction compensation method of the application, when the current mover speed v r is greater than 0, the switch s1 is turned on to realize the collection of the positive zero-crossing position error peak value e max (j).

[0009] According to the linear motor nonlinear Coulomb friction compensation method of the application, when the current mover speed v r is less than 0, the switch s2 is turned on to realize the collection of the negative zero-crossing position error peak value e min (j).

[0010] According to the linear motor nonlinear Coulomb friction compensation method of the application, the hysteresis comparators are used to compare the zero-crossing position error peak value with the corresponding zero-crossing position peak reference value to determine whether the feedback controller is turned on.

[0011] According to the linear motor nonlinear Coulomb friction compensation method of the application, whether the two feedback controllers are turned on is realized by the comparison results of the two hysteresis comparators through the switch s3.

[0012] According to the linear motor nonlinear Coulomb friction compensation method of the application, the motion cycle is the cycle of the reciprocating trajectory.

[0013] According to the linear motor nonlinear Coulomb friction compensation method of the application, the comparison results of the two hysteresis comparators are input to the AND gate, and if the comparison results of the two hysteresis comparators are within the corresponding position peak reference value, the output y1 of the AND gate is 1 to control the feedback controller to be disconnected.

[0014] The method of the application can cut in and cut out the feedback controller in the self-learning feedforward compensator to compensate the nonlinear Coulomb friction by comparing the zero-crossing position error peak value with the reference value in the dynamic adjustment process.

[0015] The method solves the problem of excessive position error peak value caused by change of the direction of Coulomb friction. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 is a flow chart of the self-learning feedforward compensator of the linear motor nonlinear Coulomb friction compensation method according to the present application; e in the figure is the current position error;

[0017] Figure 2 is a corresponding variable experimental waveform diagram of the nonlinear Coulomb friction compensated by the method according to the present application;

[0018] Figure 3 is a waveform diagram of the mover speed and the corresponding position error. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0020] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0021] The present application will be further described below in combination with the drawings and specific embodiments, but is not limited by the present application.

[0022] Specific embodiment one, in combination with Figure 1 the present application provides a linear motor nonlinear Coulomb friction compensation method, comprising,

[0023] obtaining the positive zero-crossing position error peak value e max (j) of the linear motor in each motion cycle by sampling the self-learning feedforward compensator min (j), the negative zero-crossing position error peak value e max (j) of the linear motor in each motion cycle by sampling the self-learning feedforward compensator r (j) is greater than the positive zero-crossing position peak reference value e min (j) is greater than the negative zero-crossing position peak reference value-e r (j), the negative zero-crossing position error peak value e r (j) or-e max (j) or-e r (j) or-e minThe result of (j) is used to control the output of the corresponding feedback controller to use the compensation square wave amplitude limit value for Coulomb friction compensation as feedforward compensation, which is fed forward to the current command to complete the nonlinear Coulomb friction compensation, where j represents the motion period.

[0024] Because numerous factors influence the nonlinear friction force during the velocity reversal phase, it is difficult to compensate for different reciprocating motion trajectories using a single friction model. This implementation only considers the impact of friction force on position control error during the PMLSM reciprocating motion, and based on this, proposes a self-learning feedforward compensator, such as... Figure 1 As shown in the figure. Here, y1 is an intermediate variable.

[0025] Furthermore, combined with Figure 1 As shown, when the velocity crosses zero in the positive direction, the peak position error e max (j) Peak reference value e at the positive zero-crossing point r Within, and the peak position error e at the reverse zero-crossing point of the velocity. min (j) Peak reference value at the reverse zero-crossing point -e r When the amplitude is within the range, the feedback controller is disconnected so that the compensation square wave, which serves as feedforward compensation, maintains its current amplitude output and is fed forward to the current command to complete the compensation of nonlinear Coulomb friction.

[0026] At the current mover velocity v r When the value is greater than 0, switch S1 is turned on to realize the peak position error e at the positive zero-crossing point of the velocity. max (j) collection.

[0027] At the current mover velocity v r When the value is less than 0, switch S2 is turned on to realize the peak position error e at the reverse zero-crossing point of the velocity. min (j) collection.

[0028] Furthermore, combining Figure 1 As shown, a hysteresis comparator is used to compare the peak value of the zero-crossing position error with the corresponding reference value of the zero-crossing position peak value to determine whether the feedback controller is turned on.

[0029] Whether the two feedback controllers are turned on or off is controlled by the switch s3, which is controlled by the comparison result of the two hysteresis comparators.

[0030] The comparison results of the two hysteresis comparators are both input to the AND gate. If the comparison results of the two hysteresis comparators are both within the corresponding peak reference value, then the output of the AND gate is y1=1, and the control feedback controller is disconnected.

[0031] In the embodiment, the self-learning feedforward compensator mainly comprises a hysteresis comparator, a controller and a limiting section. The sampling section collects the position error peak value in each motion cycle, including the error peak value of the positive zero-crossing of the speed and the error peak value of the negative zero-crossing of the speed. The hysteresis comparator realizes the start and stop of the controller by setting a comparison threshold, that is, when the error peak values of the positive and negative zero-crossing reach the threshold range, the controller stops working; and when the error peak value exceeds the error range, the controller starts working. The limiting section mainly limits the saturated output of the controller to ensure the safe operation of the linear motor.

[0032] The self-learning feedforward compensator samples the position error peak value in the speed zero-crossing region in each repetitive motion cycle, and obtains the position peak error e max (j) of the positive zero-crossing of the speed in the jth motion cycle and the position peak error e min (j) of the negative zero-crossing of the speed. The peak error reference values of the positive and negative zero-crossing stages are set as e r and -e r respectively. The feedback controller is used to adjust the output compensation square wave limiting values C1(j+1) and C2(j+1), and then the C1(j+1) and C2(j+1) are fed forward to the current command to complete the compensation of the Coulomb friction. The motion cycle of the feedback control is the cycle of the reciprocating trajectory.

[0033] In the dynamic adjustment process, with the decrease of the position error in the zero-crossing region, the controller in the self-learning feedforward compensator is turned on and turned off by setting the hysteresis comparator. When the position error reaches the range of the error threshold, the controller stops working, and the feedforward compensation square wave maintains the current amplitude output. However, in the subsequent motion process, the position control error may exceed the error threshold again due to the time-varying characteristics of the friction, and then the controller will start working again to adjust the amplitude of the feedforward output square wave.

[0034] Experimental verification: a linear motor experimental platform is built, and the experimental verification waveform is as shown in Figure 2 Fig. (a) is the feedback speed, (b) is the position error, and (c) is the output of the self-learning feedforward compensator compensation square wave amplitude. It can be seen from Figure 2 that with the increase of the motion cycle, the self-learning feedforward compensator can gradually converge, and the position error in the speed zero-crossing region gradually decreases, which indicates the effectiveness of the method.

[0035] While the application has been described with reference to particular embodiments thereof, it is to be understood that these embodiments are merely illustrative of the principles and applications of the present application. It will be apparent to those skilled in the art that numerous modifications can be made within the scope of the present application as defined by the appended claims. It is intended that all such modification fall within the spirit and scope of the present application. It will be understood that the features described in connection with one embodiment can be used in connection with another embodiment.

Claims

1. A method of compensating for nonlinear Coulomb friction in a linear motor, characterized by Comprising, The peak position error e at the positive zero-crossing point of the speed in each motion cycle of the linear motor is obtained by sampling through a self-learning feedforward compensator. max (j) Peak position error e at the zero-crossing point opposite to the velocity min (j), the peak position error e at the positive zero-crossing point of velocity. max (j) Exceeds the peak reference value e at the positive zero-crossing position r Or the peak position error e at the zero-crossing point of the velocity in the opposite direction min (j) Exceeds the peak reference value at the reverse zero-crossing position -e r At that time, according to e r -e max (j) or -e r -e min The result of (j) is used to control the output of the corresponding feedback controller to use the compensation square wave amplitude limit value for Coulomb friction compensation as feedforward compensation, which is fed forward to the current command to complete the nonlinear Coulomb friction compensation, where j represents the motion period.

2. The linear motor nonlinear Coulomb friction force compensation method according to claim 1, characterized in that, when the speed is positive, the peak value of the zero-crossing position error e max (j) is within the peak value reference e of the positive zero-crossing position r when the speed is negative, the peak value of the zero-crossing position error e min (j) is within the peak value reference -e of the negative zero-crossing position r when the speed is negative, the peak value of the zero-crossing position error e min (j) is within the peak value reference -e of the negative zero-crossing position r when the speed is negative, the peak value of the zero-crossing position error e min (j) is within the peak value reference -e of the negative zero-crossing position r when the speed is negative, the peak value of the zero-crossing position error e min (j) is within the peak value reference -e of the negative zero-crossing position r when the speed is negative, the peak value of the zero-crossing position error e min (j) is within the peak value reference -e of the negative zero-crossing position <000 3. The linear motor nonlinear Coulomb friction force compensation method according to claim 2, characterized in that, At the current mover speed v r When greater than 0, switch s1 is turned on, realizing the peak value e max (j) is collected.

4. The linear motor nonlinear Coulomb friction force compensation method according to claim 3, characterized in that, At the current mover speed v r When the value of e is less than 0, switch s2 is turned on, and the peak value e of the position error of the speed reversal zero-crossing point is realized min (j) is collected.

5. The linear motor nonlinear Coulomb friction force compensation method according to claim 4, characterized in that, The hysteresis comparator is used to compare the zero-crossing position error peak value with the corresponding zero-crossing position peak reference value to determine whether the feedback controller is started.

6. The linear motor nonlinear Coulomb friction force compensation method according to claim 5, characterized in that, Whether the two feedback controllers are started is realized by the comparison results of the two hysteresis comparators and the switch s3.

7. The linear motor nonlinear Coulomb friction force compensation method according to claim 6, characterized in that, The motion period is the period of the reciprocating motion trajectory.

8. The linear motor nonlinear Coulomb friction force compensation method according to claim 7, characterized in that, The comparison results of the two hysteresis comparators are input to an AND gate, and if the comparison results of the two hysteresis comparators are both within the corresponding position peak reference value, the output y1 of the AND gate is 1, and the feedback controller is disconnected.

Images